Skip to main content
BMJ Case Reports logoLink to BMJ Case Reports
. 2022 Mar 29;15(3):e248473. doi: 10.1136/bcr-2021-248473

Sinogenic polymicrobial anaerobic orbital cellulitis: the importance of source control and broad empirical antibiotics

Andrew Blunsum 1, Clare Treharne 2, Theofano Tikka 3, Arunachalam Iyer 2, Elan Micha Tsarfati 4,
PMCID: PMC8966534  PMID: 35351758

Abstract

A man in his twenties with a history of recurrent sinusitis was urgently referred to the emergency department (ED) by an out-of-hours general practitioner following a 2-day history of increasing right eye pain, redness and swelling after a week of coryzal symptoms. He denied visual impairment and any history of recent dental pain or procedures. Initial assessment in ED noted fever, tachycardia and hypotension. Video consultation with ophthalmologist in the ED identified proptosis, periorbital erythema and chemosis with full eye movement solely affecting the right eye. Visual acuity of 6/6 was confirmed in both eyes. After review by the ear, nose and throat (ENT) team, a diagnosis of sinogenic right orbital cellulitis was made, empirical antibiotics started and care transferred to the ENT team for immediate surgical intervention. 48 hours postoperatively, the patient acutely deteriorated, developing ophthalmoplegia and visual acuity of 6/95 in the right eye. Repeat imaging demonstrated a deteriorating picture and urgent surgery was organised at a neighbouring hospital’s specialist ENT unit combined with a change to his antibiotics. On day 4, 1 day following transfer, an anaerobic bacterium, Eggerthia catenaformis, was isolated from blood cultures collected on admission. The patient improved clinically following the second surgery and targeted antimicrobial therapy, eventually being discharged 10 days after initial presentation. In addition to E. catenaformis, the Anaerobic Reference Unit (Cardiff) identified two further anaerobic bacteria, Parvimonas micra and Dialister pneumosintes. This paper presents the first documented case of polymicrobial anaerobic orbital cellulitis secondary to acute bacterial sinusitis. Moreover, this case underpins the importance of broad empirical antibiotics coupled with surgical source control to effectively manage a rare but sight-threatening and life-threatening disease.

Keywords: Ear, nose and throat; Infections; Eye; Ear, nose and throat/otolaryngology; Infectious diseases

Background

Orbital cellulitis is both a sight-threatening and life-threatening disease requiring immediate antimicrobial therapy and definitive surgical management. This case is the first documented case of polymicrobial anaerobic orbital cellulitis secondary to a bacterial sinus infection without dental extension involving Eggerthia catenaformis in a young male patient requiring two surgeries to achieve source control.

Case presentation

A man in his mid-20s with a history of chronic sinusitis presented to an out-of-hours general practitioner (GP) describing an acute worsening of right-sided maxillary and frontal sinusitis symptoms. Further history revealed increasing right eye pain, redness and swelling with an acute deterioration over the previous 4 days, all reported to have acutely worsened along with vomiting on the day of seeking medical help. Immediately prior to presentation, he described a 1-week history of coryzal symptoms and denied any recent antibiotics, dental pain, history of recent dental procedures or facial trauma. He had no history of diabetes or malignancy and denied any steroid use.

On initial examination, the GP recorded a temperature of 38.4°C, heart rate of 130 beats/min, blood pressure of 99/69 mm Hg and saturations of 98% on room air. On examination, the right eyelid was swollen with a proptosis. Right eye movements reportedly elicited pain though visual acuity was unchanged. There were no abnormalities of the left eye from history or examination and all other organ systems were normal. The GP organised urgent admission to hospital for intravenous antibiotics and video review with ophthalmologist after consulting the on-call ophthalmology consultant.

On arrival in the ED, the patient reported worsening of right-sided periorbital swelling and eye pain. Right eye movements became more painful and vomiting continued. His temperature had risen to 38.8°C and he remained tachycardic. On further examination, the patient demonstrated a progressive right-sided proptosis with significant periorbital erythema and chemosis. Eye movements were painful but not restricted, and eye opening was further limited due to periorbital swelling. He had normal visual acuity of 6/6 bilaterally and his pupils were equal and reactive.

The ophthalmology reviewed the patient remotely using a video platform and a diagnosis of right orbital cellulitis was made. Intravenous co-amoxiclav 1.2 g three times daily was started immediately after blood cultures were collected. Additionally, analgesia, antiemetic, intravenous fluids and xylometazoline 0.1% nasal drops 2 drops in every 6 hours were commenced.

An urgent ear, nose and throat (ENT) referral was made for immediate inpatient care and prompt consideration of incision and drainage of a presumed orbital abscess. Given the rapidly progressive features of the eye swelling and concerns for rising intraocular pressure, a single dose of acetazolamide 500 mg intravenous was given and urgent CT imaging of the head was arranged.

Investigations

Biochemistry and haematology

Results from the patient’s inpatient stay are listed in table 1.

Table 1.

Haematology and CRP

Hospital 1 Hospital 2
ED/admission Day 2 Day 3 Day 4 Day 6 Day 8
Haemoglobin
(130–170 g/L)
147 133 135 139 136 145
White blood cells
(4×109/L–10×109/L)
12.0 20.5 21.1 14.3 18.8 18.1
 Neutrophils
 (2.0×109/L–
 7.0×109/L)
11.3 19.2 17.0 11.5 15.2 13.6
 Lymphocytes
 (1.0×109/L–
 3.0×109/L)
0.4 0.8 2.5 2 3 3.7
CRP
(<5 mg/L)
167 321 99 60 8 <6

Bold values are abnormal.

CRP, C reactive protein; ED, emergency department.

Imaging

On the day of admission, prior to surgery, a contrast CT of the brain and orbits (figure 1) demonstrated extraconal, retro-orbital gas locules bordering the medial right orbital wall with associated fat stranding but no fluid component to suggest definite abscess formation. Marked ipsilateral preseptal soft tissue thickening extended into the medial canthus. Complete opacification of the right maxillary sinus was noted with partial opacification of the right frontal and ethmoid sinuses, without breach of the lamina papyracea. There was no intracranial or extra-axial abnormality and normal venous sinuses drainage. No dental extension was identified.

Figure 1.

Figure 1

CT of the head and orbits and paranasal sinuses at time of initial presentation showing significant right-sided proptosis, preseptal oedema, gas locules along right medial orbital wall and opacified anterior ethmoid air cells (left image: axial plane; right image: coronal plane).

Differential diagnosis

Orbital cellulitis has several possible preceding causes, of which the two most commonly identified are sinogenic invasion from an acutely infected maxillary sinus and odontogenic-related infection.1

Acute bacterial sinusitis is an uncommon complication of a usually self-limiting viral sinusitis. In such cases, bacteria commonly found in the upper respiratory tract can infect one or many sinuses. When identified early, this can be treated with empirical oral antibiotic therapy in the community. Rarely the infection can progress rapidly through the sinuses via fissures or breaks in the sinuses into the orbital space and develop orbital cellulitis. Patients often present with a history of coryza and sinusitis that has acutely worsened over 8–10 days.

Odontogenic orbital cellulitis may start from any tooth, but most commonly begin at maxillary root apices.2 3 These are anatomically adjacent to the maxillary sinus from which oral bacteria spread, often aggressively, through the sinuses and into the orbit region via any available defect or fissure present. Patients often have a history of poor dentition, recent dental infection or dental treatment.

In much rarer circumstances, acute sinusitis can be fungal in origin. These infections are commonly caused by moulds and are commonly seen in immunosuppressed patient groups, though higher rates are observed in some geographical settings.4

A diagnosis of orbital cellulitis secondary to acute bacterial sinusitis was jointly made by the ophthalmologist and ENT consultant based on the patient’s history of coryza with sinusitis without dental involvement, deranged vital signs and classical clinical findings of proptosis, chemosis and pain on eye movement and supported by imaging (figure 1).

Treatment

Surgery

In view of the initial clinical and radiological findings, a decision was taken to take the patient to theatre for immediate surgical management on the day of admission. The patient underwent right-sided endoscopic uncinectomy, middle meatal antrostomy, right anterior ethmoidectomy and opening of the lamina papyracea. Pus was found on exploration of the right-sided nasal sinuses, which was evacuated and swabbed for culture. External exploration of the eye was also attempted with a small superiorly based Lynch-Howarth incision, but no abscess was identified. A lateral canthotomy reduced intraocular pressure in the right eye from 60 mm Hg to 40 mm Hg.

Postoperative period

Despite the initial procedure, the patient’s symptoms persisted over the next 24 hours with an increase noted in his inflammatory markers on day 2 (table 1). He remained haemodynamically stable and afebrile.

On day 3, the patient acutely deteriorated complaining of right-sided headaches and worsening pain on right eye movement. On examination, he had marked proptosis and chemosis, newly reduced periorbital and corneal sensation and had developed new and significant ophthalmoplegia in all directions. He had markedly reduced visual acuity (6/95) in the right eye. There was no relative afferent pupillary defect. His white cell count had also increased (day 3, table 1). The blood culture collected at admission had come positive on the morning of day 3 with gram-positive cocci, which appeared to be in chains from the anaerobic bottle only.

Differential diagnoses for acute postoperative deterioration

The acute postoperative deterioration may have been due to one or a combination of factors, including cavernous venous sinus thrombosis (CVST), inadequate antimicrobial treatment, central nervous system extension or incomplete source control.

CVST is a serious complication of acute bacterial sinusitis requiring early intervention.5 Early signs of CVST include headache, fever, ptosis, chemosis, fever and ophthalmoplegia. Left untreated it can progress to papilloedema and extend to involve adjacent structures, including the other eye, neighbouring venous sinuses and the brain. Diagnosis is made by CT venogram and treatment involves antimicrobial therapy and heparin-based anticoagulation. Surgery may also be indicated to remove any infected material driving CVST.

Inadequate antimicrobial coverage could account for the sudden deterioration, particularly in the context of a newly positive blood culture with Streptococci present. Organisms inhabiting the respiratory tract can contain toxin-mediated virulence factors such as those harboured by Streptococcus pyogenes. They may also be co-amoxiclav resistant, including methicillin resistant Staphylococcus aureus (MRSA). Antibiotic agents such as clindamycin and linezolid target the 50S subunit of gram-positive organisms responsible for toxin production and also treat MRSA.6 7 Gram-negative organisms such as Haemophilus influenzae may also harbour resistance to co-amoxiclav while remaining susceptible to ceftriaxone.

Penetration of antibiotics into the tissues was an equal consideration. Relative to co-amoxiclav, third-generation cephalosporins such as ceftriaxone penetrate well into the cerebral spinal fluid, where the meninges are inflamed, brain abscesses and into extravascular spaces such as the sinuses. Further cultures, including blood cultures and theatre pus samples, would enable a broader understanding of the causative organism(s) and any resistance mechanisms present.

Finally, incomplete source control would lead to the acute deterioration. The anatomy of the skull leaves little space for tissues to expand so small increases in pus, gas or inflammation can lead to disastrous consequences of the venous, arterial and nerves supplying the region. Urgent incision and drainage are indicated in such situations.

To investigate these differentials a CT of the head with venogram was requested urgently. The newly positive blood cultures with Streptococci raised the possibility of a toxin producing S. pyogenes being involved. Antibiotics were escalated to intravenous ceftriaxone 2 g in every 12 hours, intravenous linezolid 600 mg in every 12 hours and intravenous metronidazole 500 mg in every 8 hours. The switch from co-amoxiclav to ceftriaxone was to target aerobic gram-positive and gram-negative upper respiratory tract organisms, particularly if these had breached the central nervous system. Finally, oral linezolid 600 mg two times per day was added for broad gram-positive cover (including MRSA) as well as for its toxin-mediating effect, and to provide additional broad anaerobic cover using a different mode of action to metronidazole.7

Further investigations

Imaging

A non-contrast CT scan of the brain, orbits and sinuses demonstrated persistent and worsening right-sided retro-orbital extraconal gas locules along the medial orbit wall and slightly more prominent tissue swelling extending posteriorly, representing an evolving abscess. The right orbital preseptal swelling was also worse with new gas formation noted (figure 2). CT venogram undertaken simultaneously excluded cavernous sinus thrombosis.

Figure 2.

Figure 2

CT of the head and orbits and paranasal sinuses: at 48 hours post initial surgical intervention showing worsening proptosis, preseptal gas formation and persistent gas locules along medial orbital wall and tissue thickening with fat stranding (left image: axial plane; right image: coronal plane).

Microbiological investigations

A blood culture set taken on admission, incubated in the BacT/ALERT Virtuo blood culture system (bioMérieux), isolated gram-positive cocci in chains in the anaerobic culture bottle 48 hours into the patient’s admission. After 1 day of incubation, a faint growth was noted exclusively on the anaerobic agar (fastidious anaerobic agar with horse blood; FHB). A gram stain performed on the growth identified non-sporing gram-positive bacilli. An identification of E. catenaformis was made by Matrix-Associated Laser Desorption Ionised Time-Of-Flight Mass Spectrometry (bioMérieux VITEK MS) on day 4 of admission. Susceptibility testing of co-amoxiclav, clindamycin, penicillin and metronidazole was performed using antibiotic-containing gradient strips on Muller-Hinton blood agar incubated in anaerobic conditions. The organism was sensitive to the four aforementioned agents.8

As the organism identified was unusual and did not match the gram stain, a sample of the blood culture fluid was sent to the Anaerobic Reference Unit (ARU; Public Health Wales) to confirm the identification and sensitivity profile.

Swabs of pus collected intraoperatively on both first and second procedures failed to isolate any growth despite incubation on a variety of media in aerobic and anaerobic environments. No gram stain was performed on the specimens due to a pre-laboratory labelling error at the time of ordering. This error made the theatre specimens appear as though they emanated from a ward environment, where such an examination would not be performed without prior discussion.

Tests for syphilis immunoglobulin G (IgG)/IgM, HIV antibody, hepatitis B surface antigen and hepatitis C antibody were negative.

Treatment

The repeat CT excluded CVST but confirmed the suspicion of progressive right orbital abscess (figure 2). In response to this oral acetazolamide 250 mg four times per day and intravenous dexamethasone 6.6 mg in every 12 hours were initiated to reduce periorbital oedema and ocular pressure.

The patient was transferred immediately to a neighbouring specialist ENT unit for further surgical management.

Second surgical procedure

Figure 3 shows the eye appearance at the outset of the second procedure. The patient underwent a revision right external incision and drainage using a full Lynch-Howarth incision and endoscopic sinus surgery. Two subperiosteal abscess pockets were identified—one superiorly behind the level of anterior ethmoidal artery and another mid-inferiorly—and drained with two swabs sent for culture. The lamina papyracea remnants from the first procedure were removed. The eye pressure and swelling significantly improved at this stage. Incision of the orbital periosteum with a phaco knife revealed a further orbital abscess which was drained (figure 4). A corrugated drain was left in place at the incision site. Revision endoscopic sinus surgery was then performed for further clearance of the right maxillary and anterior ethmoid sinuses.

Figure 3.

Figure 3

Right eye appearances during revision surgery, showing marked periorbital oedema and chemosis.

Figure 4.

Figure 4

Left image: intraoperative appearances of superior subperiosteal abscess. Right image: inferior subperiosteal abscess and white colour changes in the periorbita and orbital abscess with fat prolapsing through incision.

Postoperative management

The patient completed a 4-day course of intravenous dexamethasone 6.6 mg in addition to topical chloramphenicol eye ointment to aid healing at the area of the lateral canthotomy. Xylometazoline nasal spray and steroid nasal spray to the right nostril were prescribed as management of acute sinusitis.

Outcome and follow-up

The patient improved following two surgical evacuations of the orbital abscesses and increasingly targeted antibiotic therapy. The corrugated drain was removed after 48 hours once oedema had significantly regressed. Ophthalmology review in the immediate postoperative period revealed significantly reduced proptosis and improvement in eye movements. Visual acuity in the right eye had also improved and the patient was able to appropriately count fingers at 30 cm. On the final ophthalmology assessment, the visual acuity was 6/9 on the right eye. There was minimal periorbital bruising with small amount of subconjuctival haemorrhage inferiorly. A notch was visible on the right lower eye lid at the region of the previous canthotomy. There was no residual ophthalmoplegia, pupils were equal and reactive, and funduscopy revealed no abnormalities. By day 8, the patient’s CRP had completely normalised (table 1).

The patient completed 10 days of intravenous ceftriaxone and metronidazole. None of the swabs collected in theatre yielded any growth. Following sustained improvement and final sensitivity results from the local laboratory for the E. catenaformis, linezolid was stopped on day 8. On hospital discharge, antibiotics were switched to oral co-amoxiclav 625 mg in every 8 hours for a further 7-day course. In total, the patient completed 17 days of antibiotic therapy.

Two weeks post second procedure, the ARU confirmed the identification of E. catenaformis along with Parvimonas micra and Dialister pneumosintes. All three organisms were reported as sensitive to penicillin, co-amoxiclav, metronidazole, clindamycin and meropenem. The E. catenaformis and P. micra were additionally sensitive to vancomycin.

Approximately 12 months later, the patient’s vision returned to normal and the incision scar healed well (figure 5).

Figure 5.

Figure 5

Recovered. A selfie showing the scar of the incision on the medial aspect of the right eye approximately 12 months later.

Discussion

The aetiology of orbital cellulitis often results from extension of acute bacterial sinusitis.2 Although it can affect all age groups, it is more common within the paediatric population.3 Prompt treatment involving surgery and antibiotics is required to prevent severe complications, including loss of vision, CVST, brain abscess and death.1 A radiological classification system proposed by Chandler and colleagues is used to describe the severity of orbital complications secondary to sinusitis.3 These are separated into five groups outlined in table 2.

Table 2.

Chandler’s classification

Group I Preseptal cellulitis: inflammation and oedema anterior to orbital septum
Group II Orbital cellulitis: extension of inflammation to include orbital contents posterior to septum
Group III Subperiosteal abscess: development of mucopurulent collection between bony orbital walls and periorbita
Group IV Orbital abscess: development of mucopurulent collection/s within the orbital contents
Group V Cavernous sinus thrombosis

Sinogenic orbital cellulitis is frequently complicated by discrete pockets of purulent material that can be challenging to access and drain due to their position and location.9 Conventional surgical drainage is performed using an external approach such as the Lynch-Howarth incision carried out in this case.10 Endoscopic sinus surgery can also be used to perform ethmoidectomy, removing the lamina papyracea to allow abscess drainage.9 In this case, a second surgical procedure was required to release all purulent material leading to the acute deterioration this patient experienced on day 3. The patient improved rapidly once source control was achieved.

The gas formation along the extraocular muscles with an intact lamina papyracea was likely the result of the anaerobic bacteria isolated in the admission blood culture. All anaerobes produce gas to varying extents. When introduced to a favourable foreign environment, anaerobes cause destruction using a variety of virulence factors, including those which favour invasion and tissue destruction. In this case, the patient had multiple anaerobes isolated, each exerting unique virulence factors in addition to gas production resulting in the increased ocular pressure observed on day 3, justifying the need for urgent surgical intervention.

Empirical antibiotics choice

Understanding the aetiology of orbital cellulitis is an important early step in the management of this disease in order to manage the source surgically and target antimicrobial therapy most appropriately. Bacteria most commonly involved in acute bacterial sinusitis, and thus likely involved in orbital cellulitis, are S. pneumoniae, H. influenzae, S. aureus, Moraxella catarrhalis and S. pyogenes. Importantly, the first three organisms can exhibit resistance to penicillins and, less commonly, cephalosporins, so careful attention to resistance rates, previous microbiology and medical, social and travel history are relevant when initiating empirical antimicrobial therapy.11 12

Odontogenic orbital cellulitis infections are typically polymicrobial, containing aerobic and anaerobic species. Common aerobes include S. aureus, viridans-group and anginosis-group Streptococci and coagulase-negative Staphylococci.13 Anaerobes are abundant in the mouth and include Eggerthia, Dialister and Parvimonas species, among others.13–15

P. micra and D. pneumosintes species can cause a variety of infections of the head and neck, including dental abscesses and blood steam infections.14 15 P. micra is an anaerobic gram-positive coccus, which has been associated with metastatic spinal infection from odontogenic foci, which may account for 45.1% of metastatic P. micra cases.16 P. micra has also been isolated from cases of pleural effusion, infectious endocarditis, septic arthritis, renal abscess, psoas abscess, brain abscess, pulmonary abscess and meningitis.17 18 P. micra is usually susceptible to antimicrobials commonly used to treat odontogenic infection.19 D. pneumosintes is an anaerobic gram-negative rod associated with complicated periodontal infection, leading in some cases to subdural or brain abscesses, bacteraemia and sepsis.20–22 In contrast to P. micra, D. pneumosintes may exhibit resistance to metronidazole, pristinamycin and rifampicin, although susceptibility to beta-lactam antibiotics such as amoxicillin and co-amoxiclav, and lincosamides such as clindamycin, is retained.23 It is worth reiterating that our patient had no dental involvement from either the history, examination or radiological findings. It is, therefore, remarkable that solely anaerobic oral flora were recovered in this case.

Sudden deterioration

The patient’s sudden deterioration on day 3 with a blood culture gram stain newly demonstrating Streptococci gave rise to the possibility of a toxin-producing S. pyogenes. Other bacteria that can exhibit toxin-mediated effects also include Panton-Valentine Leukocidin S. aureus (including MRSA). Given a nearly 40% resistance rate to clindamycin among Staphylococci and Streptococci in our region, it was prudent to select linezolid, an agent with a low likelihood of resistance.

Linezolid has also shown activity against a broad range of gram-positive and gram-negative anaerobes, including isolates from oral flora.6 This was particularly important in this case as the switch from co-amoxiclav to ceftriaxone meant a loss of anaerobic cover. While it should not be considered on par with metronidazole or beta-lactams, linezolid exhibits activity against many clinically relevant anaerobes and can add supplementary anaerobic cover in the context of a deteriorating patient.

E. catenaformis and challenges in identifying anaerobes

In this case, E. catenaformis was the only organism identified from the blood culture by the local laboratory. E. catenaformis is a gram-positive, anaerobic, non-spore forming rod, which has previously been implicated in severe dental infections.24 It is an obligate anaerobe and grows well on FHB agar. E. catenaformis is typically sensitive to first-line agents such as penicillin, metronidazole and clindamycin as well as co-amoxiclav, piperacillin/tazobactam and meropenem. No information exists on antibiotic resistance. To the best of our knowledge, this is the first case of orbital cellulitis in the literature in which E. catenaformis has been identified as one of the causative organisms.

Isolating anaerobes from small pockets of pus can be challenging and attention to specimen collection is important. Collecting the largest quantity of available specimen into a universal container maintains the viability of anaerobic organisms, using a syringe to access small quantities if necessary. Keeping the sample away from heat and light also improves the likelihood of growth. Blood culture media are adept at culturing anaerobes from blood but can also be inoculated with pus and fluids to increase yield and take advantage of the antimicrobial neutralising media housed within.25 Swabs, as used repeatedly in this case without yielding any growth, are least favoured because the volume of pus is small, often exposes anaerobes to too much oxygen and some dry out prior to being capped. Every effort should be made to limit transport delays as recovery is greatest from specimens that are received by the microbiology laboratory within 20 min of collection and incubated on appropriate agar and in broths.

Once in the microbiology laboratory, it is vital anaerobic cultures are maintained in strictly anaerobic conditions between 35°C and 37°C for up to 5 days.25 Handling anaerobic plates outside of the anaerobic cabinet must be limited to avoid too much oxygen exposure.

It is worth noting that in this case, it remains unclear why two other organisms in the blood culture failed to grow in the local laboratory. The local procedures for handling anaerobes are robust and great care was taken with this sample. However, it does raise the importance and value of the ARU. The ARU have the skill, specialist media and expertise to identify anaerobic bacteria and readily available to discuss the case if needed. Their website and handbook outline in detail how their services operate, what samples they can analyse and how susceptibility testing is performed on anaerobes. In this case, two conversations between the microbiology consultant and the ARU proved helpful ensuring the local identification and sensitivity pattern of the E. catenaformis were correct. The additional identification and sensitivity profile of the two other anaerobes was a great help as it confirmed the initial gram stain was not an error but that the P. micra and that D. pneumosintes were also party to the infection.

Patient’s perspective.

First night

I had a perceived sinus infection where symptoms became progressively worse as the day went on. I started to feel a further buildup of pressure in my sinuses around 20:00 causing discomfort and dizziness. Not long after I decided to phone NHS24 and told they would phone back. Around 21:00, I started to vomit uncontrollably as NHS phoned back, my girlfriend had to take the call. We then were sent to the out of hours clinic (OOHC) around 22:30.

Clinic

At the OOHC, the vomiting continued and the dizziness got worse. The clinic staff were taking questions as the threat of COVID-19 had only just started. During the questions, I had to lie down, my right eye began to swell, and my skin tone started to go grey. I was told to go to the A&E after she could not diagnose what was wrong and I became more out of it.

Hospital

I arrived to the A&E around 23:30 and doctors gave me something for the vomiting. I met the ophthalmologist consultant over a video call while A&E staff were doing eye tests to further diagnose the symptoms. I was admitted into a ward around 04:30.

The following day is a bit blurry. The doctors were doing test before the operation, a CT scan at 15:00 and then admitted for the first surgery around 18:00. The pressure had been relieved due to a drain being put in and gas and pus being released. I was fine for the day proceeding that. Then morning after, the pressure built back up again. I started to become severely dizzy again and they then rushed me to the nearby specialist centre by ambulance.

Specialist centre

The second surgery took place as soon as I arrived which was successful. I was then kept in for about 2 weeks for daily check-ups, antibiotics and supervision.

Recovery

Once discharged, I was left with an open incision at the right side of my eye and some stitches. I stayed connected with ophthalmology and I took further oral antibiotics. Once the stitches were out, there were no follow-up appointments or check-ups.

Reflection

My eye healed though there are scars where the incisions were (figure 5). I am thankful it was not a worse outcome. A massive thank you to the staff at both hospitals, I was well looked after at a very uncertain time for everyone. The value of the NHS is undeniable. The worst thing about being in that situation was the unknown aspects, when the doctors are unsure, you know you are in for a bumpy ride. All of that on top of COVID-19’s grasp on the country made for quite a scary cocktail, lucky there were some prescription drugs and Netflix in the mix.

Learning points.

  • Definitive source control, often with open surgery, must be achieved in orbital cellulitis to prevent permanent sight loss and death.

  • Anaerobic bacteria are a well-documented contributor in odontogenic orbital cellulitis, though this is the first documented instance of a polymicrobial anaerobic orbital cellulitis secondary to acute bacterial sinusitis.

  • Eggerthia catenaformis is an anaerobe not previously described in association with acute bacterial sinusitis or orbital cellulitis.

  • In addition to anaerobic cover, empirical antibiotic regimens for acute bacterial sinusitis extending into the orbital space must include cover for Haemophilus influenzae, Staphylococcus aureus, Streptococcus pneumoniae, beta-haemolytic Streptococci and viridans-group Streptococci, taking into account previous microbiology results, local resistance rates, medical, social and travel history.

  • Recovery of anaerobes in many laboratories is challenging due to preanalytical factors and the fastidious nature of these organisms, even when appropriate anaerobic media is used in a strictly controlled anaerobic environment. Involvement of a reference laboratory is vital, even prior to collecting cultures, where anaerobes are suspected to discuss treatment and sampling.

Acknowledgments

We wish to acknowledge and thank the UK Anaerobe Reference Unit (Cardiff, UK) for their skill and expertise in this case. They are perennially helpful and the entire laboratory team did incredible work. Many thanks to the NHS FV microbiology laboratory as well for their incredible effort in identifying the unusual anaerobe in this case.

Footnotes

Contributors: ET: have been involved in the conception, drafting (writing the microbiology/antimicrobial related sections) and have edited/revised/proofread the manuscript, also act as the final approver and will remain accountable for the final article, and also the corresponding author. AB: wrote most sections of the manuscript, including developing tables, editing and proofreading. CT and TT: involved in writing the manuscript, specifically sections focused on ENT management. AI: involved in editing and providing clinical images.

Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

Case reports provide a valuable learning resource for the scientific community and can indicate areas of interest for future research. They should not be used in isolation to guide treatment choices or public health policy.

Competing interests: None declared.

Provenance and peer review: Not commissioned; externally peer reviewed.

Ethics statements

Patient consent for publication

Consent obtained directly from patient(s).

References


Articles from BMJ Case Reports are provided here courtesy of BMJ Publishing Group

RESOURCES